Abstract CT209: Correlation of robust local reactions prompting GM-CSF dose reduction to clinical response in a phase II trial of the AE37+GM-CSF HER2 peptide vaccine

Abstract

Abstract Background: We are conducting a phase II clinical trial of the HER2 peptide vaccine AE37+GM-CSF for prevention of breast cancer recurrence in disease-free, node-positive or high-risk node-negative patients (pts), who have completed standard of care therapy. AE37, an Ii-Key hybrid of the HER2-derived peptide AE36 (aa:776-790), is an MHC Class II epitope capable of stimulating tumor reactive CD4+ helper T-cells. Previous investigation of intradermal inoculation with E75 (HER2 aa:369-377) +GM-CSF vaccine in phase I/II trials showed no recurrences at 24 months among the 18% of pts requiring a GM-CSF dose reduction due to robust local reactions (LR). Phase I/II studies have shown the safety, immunogenicity, and potential clinical benefit of AE37 + GM-CSF. Here, we examine the relationship between GM-CSF dose reduction and clinical recurrence rate (RR) in pts receiving AE37+GM-CSF. Methods: Patients with any level of HER2 expression (IHC1-3+) were enrolled and randomized to receive 6 monthly intradermal inoculations of AE37+GM-CSF or GM-CSF alone (controls) during the primary vaccination series, then four booster vaccinations administered every 6 months. Enrollment has been completed and pts continue with their vaccination series. LRs are measured after each inoculation, and pts with a LR measuring ≥ 100 × 100mm have their GM-CSF dose reduced on subsequent inoculations. Local and systemic toxicity are being monitored and clinical recurrences documented. Proportional RR are compared using a chi-square or Fisher exact test as appropriate. Results: Of 301 enrolled pts, 154 were randomized to the vaccinated arm (VG) and 147 randomized to the control arm (CG). The groups are well-matched for clinicopathologic characteristics. Toxicities have been almost exclusively grade 1 and 2. Study-wide, 18.9% of pts required dose reduction (CG 15.6%, VG 22.1%; p = 0.19). In vaccinated pts, the RR in the dose reduced group (DR) was 5.9% (2/34) versus the non-dose reduced group (NDR) RR of 14.2% (17/120) (p = 0.25). Among controls, the DR RR was 17.4% (4/23) vs the NDR RR of 12.9% (16/124) (p = 0.74). Comparing all DR pts, the relative risk of recurrence was reduced by 66% in the VG DR compared to the CG DR (5.9% versus 17.4%; p = 0.21). Conclusions: In a randomized phase II trial of the AE37+GM-CSF vaccine, pts who required a dose reduction of GM-CSF due to robust LR trend toward a lower recurrence rate. The overall dose reduction rate is similar to that in E75 trials (18.9% vs 17.9%) and occurred more frequently in the VG than CG. While dose reduction is seen in both the VG and the CG, the clinical benefit is only seen in the VG suggesting that while the LR may be a result of the GM-CSF, the specificity provided by the peptide is required for clinical benefit. Furthermore, these data suggest that GM-CSF should be dosed to produce large LR to enhance the benefit of peptide vaccines in the adjuvant setting. Citation Format: Julia M. Greene, Erika J. Schneble, Jennifer K. Litton, Jonathon Martin, Alfred F. Trappey, John S. Berry, Timothy J. Vreeland, Diane F. Hale, Guy T. Clifton, Alexandros Ardavannis, Michael Papamichail, Sonia Perez, Sathibalan Ponniah, Elizabeth A. Mittendorf, George E. Peoples. Correlation of robust local reactions prompting GM-CSF dose reduction to clinical response in a phase II trial of the AE37+GM-CSF HER2 peptide vaccine. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr CT209. doi:10.1158/1538-7445.AM2015-CT209